JP2002218452A - Method and device for monitoring moving object - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for detecting a relatively slow moving object within a video where a large quantity of noise that fast changes occurs. SOLUTION: Photographed images of a target area are sequentially stored as much as the prescribed number of images in equal space with the latest photographed image stored first, a first difference image DI1 composed of the difference between the stored latest photographed image It and its adjacent photographed image It-1 and a second difference image DIn composed of the difference between the latest photographed image It and a photographed image It-n away in prescribed space are formed, and a composite image SIt is formed to be defined as an intermediate image by performing image subtraction of the first difference image DI1 from an image obtained by multiplying an image obtained by performing image addition of a previously calculated intermediate image SIt-1 to the second difference image DIn by an attenuation coefficient (a). A binarized image BIt is subsequently obtained by binarizing the calculated composite image SIt by using a prescribed threshold during repeating the same process, and when a region being larger than a prescribed area is detected from the video of the binarized image, the region is reported.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for automatically determining a moving object and monitoring a moving state, and more particularly to a monitoring method and apparatus for automatically detecting a ship moving on water.

[0002]

2. Description of the Related Art Commercial vessels such as freighters and fishing boats often come and go, especially in harbors, near seas, and barriers.
It is necessary to monitor the sailing vessels from the viewpoint of maritime traffic safety. For example, port managers use both visual and TV cameras to monitor vessels entering and leaving the port, but many vessels sail outside of the regular time, and make sure that there is no drop during times when the number of vessels is extremely small. Monitoring is necessary, so the load on the administrator is large.

[0003] In order to reduce the burden on the administrator, it would be convenient if a monitoring device that automatically detects and alerts a navigating ship can be used. As a method of recognizing a moving object using a captured image, a method of creating a background image in which no moving object exists and detecting the moving object by taking a difference from a measurement image, or a method of photographing at a fixed time interval There is known a method of detecting and displaying a moved portion in an image by calculating a difference between the images. However, the method using a background image works effectively in an environment with a steady background, such as indoors, but it is not useful in an environment where the sunshine changes or the background scenery changes with time. Not available to do.

On the other hand, a method of obtaining a difference between images at time intervals can be used even in an outdoor environment where there is an environmental change. As an application of this method to ship monitoring, see, for example,
At -65299, the observation area set for the water surface of the port or river was photographed with a TV camera, and the video image data captured at appropriate time intervals was binarized, and then successively subtracted for each adjacent time. Further, there is described a water surface monitoring system in which a logical product is obtained between adjacent obtained difference images to extract a portion common to the two and detect a moving body and track the position of the center of gravity.

However, in the detection method using the image difference, if there is an object whose brightness changes in the background, it becomes noise in the difference image and the moving object cannot be correctly extracted. Especially when monitoring moving objects on the water surface, the sun reflects on the surface of the water or the surface of the water undulates, causing background noise with a large brightness difference that changes at high speed in the screen. It is very difficult to extract from images. The above-mentioned publication discloses that the photographing time interval is made as short as possible in order to suppress the generation of noise. This is because if the brightness noise is captured in two consecutively shot images, it can be eliminated by canceling. On the other hand, if the time interval is shortened, it becomes difficult to catch the movement of the ship, so that it is said that the images are taken every 0.2 to 0.3 seconds as a result of considering the overlap of the images of the ship. However, when monitoring vessels over a long distance by navigating the surface of the water, such as port departure and entry control, it is difficult to detect relatively low-speed movement of commercial vessels such as commercial vessels at such an interval. is there.

[0006]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method for detecting a relatively slow moving object from an image in which a large amount of rapidly changing noise is generated. In particular, it is an object of the present invention to provide a monitoring method and a monitoring device for detecting a ship moving on the water surface.

[0007]

In order to solve the above-mentioned problems, a moving object monitoring method according to the present invention captures a photographed image of a target area at a predetermined cycle, and connects the mutually adjacent images each time the photographed image is captured. The difference between the first difference image and the image separated by a predetermined distance is taken as a second difference image, and the second difference image is added to a predetermined number of images to obtain a first difference image.
A combined image is formed by subtracting the difference image obtained by adding the same predetermined number of images to form a composite image, and a notification is made when an area larger than a predetermined area is detected from a binarized image obtained by binarizing the composite image. And

The moving object monitoring method of the present invention focuses on the fact that changes in the waves on the sea surface and the state of reflection of sunlight are sufficiently fast as compared with the moving speed of a moving object such as a ship, and uses simple image processing. The processing corresponding to the band-pass filter that has been performed cancels noise to detect a moving object. If a second difference image is obtained by taking an image difference between temporally distant images among images acquired at equal intervals, an object that changes at a low speed such as a ship can be detected. However, since the second difference image includes a fast-changing noise component such as a sea surface wave or a change in sunlight, clear movement detection cannot be performed.

On the other hand, if the first difference image is obtained by calculating an image difference between mutually adjacent images among images acquired at equal intervals, a noise component that changes at a high speed can be detected. However, since the change in the sea surface occurs quickly and the position of occurrence is not determined, the noise in the first difference image does not overlap with the noise in the second difference image, and a process of subtracting the second difference image from the first difference image is performed. This does not offset the effects of noise.

However, since noise is detected evenly in a region where a sea level change occurs, the noise generation region of the first difference image and the noise generation region of the second difference image have the same level of noise when integrated a suitable number of times. Therefore, by subtracting the integrated image of the first differential image from the integrated image of the second differential image, a noise component in a noise generation region such as the sea surface can be canceled to extract a moving object such as a ship. It is needless to say that the present method can be widely used not only for vessels navigating on the water but also for detecting noise that changes at a high speed and a moving object that changes at a low speed.

Further, in order to solve the above-mentioned problems, a second method for monitoring a moving object according to the present invention captures an image of a target area at a predetermined cycle, and sequentially captures images of the target area at equal intervals from the latest one. Store only the number. A first difference image that is a difference between the stored latest captured image and an adjacent captured image and a second difference image that is a difference between the latest captured image and a captured image separated by a predetermined interval are formed. The second difference image is added to the image and multiplied by the attenuation coefficient, and the first difference image is subtracted from the image to form a composite image to be an intermediate image.
Thereafter, while the same process is repeated, the obtained composite image is binarized using a predetermined threshold to obtain a binarized image, and a notification is made when an area larger than a predetermined area is detected from the image of the binarized image. It is characterized by doing.

According to the second moving body monitoring method of the present invention, a low speed image of a ship or the like can be obtained by using a second difference image obtained by calculating an image difference between an image obtained at equal intervals and an image temporally separated from the latest image. Detect objects that change. Here, too, a noise component that changes at a high speed is detected from the first difference image obtained by calculating the image difference between the latest image and the immediately preceding image among the images acquired at equal intervals, and the first difference image is used as the second difference image. The noise component generated in the noise generation area in the second difference image such as the sea surface is canceled by repeating the operation results an appropriate number of times and integrated, and the sea surface wave or sunlight change included in the second difference image is removed. A moving object such as a ship changing at a low speed can be detected from the background of the image.

It is preferable that the term to be integrated is multiplied by a damping coefficient to suppress the degree of influence of past events so as not to be affected by past events. The attenuation coefficient determines the width of the window, and how much past events are considered can be adjusted by the magnitude of the attenuation coefficient. In addition, the presence of a moving object can be detected by a shape in which a moving portion of a low-speed moving object, which appears as a difference between two images separated in time, overlaps for a predetermined time. The overlap of the moving parts is a figure formed at the front and rear ends of the moving body, having the width of the moving body and having a length in the moving direction. Since it is displayed as thin as possible, it can be easily recognized by a person.

In the method of the present invention, in order to further mechanically extract the moving object, the composite image is converted into a binarized image using an appropriate threshold value, and the overlap of the moving parts is extracted.
When the overlapping area has an appropriate size, it is recognized as a moving object. The threshold value used at this time may be determined by taking a density histogram of the photographed image and using the result. Noise generation conditions such as sea surface reflection are greatly affected by the environment. Under the direct light condition in which sunlight is irradiated from behind the camera, the entire screen becomes evenly bright and the screen has a small spread of the density histogram. Therefore, the threshold value is lowered so that the moving object can be accurately detected. However, under backlight conditions, high-brightness areas occur due to the reflection of sunlight on the sea surface, increasing the contrast with other areas, and the effect of sea-level changes is greater than in normal light, which reduces the detection performance of the mobile object itself. Even so, it is preferable to set a high binarization threshold to remove fine noise. Under such conditions, the width of the histogram is widened, so that the threshold can be determined using the distribution of the histogram as an index.

In order to solve the above-mentioned problems, a moving object monitoring apparatus according to the present invention is characterized in that an imaging device captures an image of a target area at a predetermined cycle, and a captured image memory stores captured images at equal intervals acquired by the imaging device. A predetermined number of images are sequentially stored from the latest image, and the image processing apparatus generates a first difference image by taking an absolute value of a difference between the latest image and an adjacent image, and calculates a difference between the latest image and an image separated by a predetermined interval. A second differential image is generated by taking an absolute value, and the intermediate image recorded in the intermediate image memory and the second differential image are image-added and multiplied by an attenuation coefficient, and the first differential image is subjected to image subtraction to obtain a composite image. Is formed, the combined image is stored in an intermediate image memory, and a binarized image obtained by binarizing the combined image is subjected to labeling processing to detect an area larger than a predetermined area. Characterized by

[0016] The binarization threshold may be determined based on a histogram relating to the lightness in the captured image. Further, it is preferable to saturate at an upper limit value or a lower limit value for each pixel of the intermediate image. Since the pixel value of the digital display type returns to the lower limit immediately after exceeding the upper limit, and returns to the upper limit when exceeding the lower limit, it is preferable to saturate in each direction so that the calculation result does not become abnormal. It is.

The moving object monitoring apparatus of the present invention implements the second moving object monitoring method of the present invention, and cancels a high-speed changing noise component such as a sea surface wave or a sunlight reflection state, and a low-speed changing moving object such as a ship. Can be detected. In addition, since the pixel value of the intermediate image obtained as a result of the image operation including the integration may exceed the physical limit, the upper limit and the lower limit are set, and when the upper limit and the upper limit are exceeded, the saturation is performed, and the abnormal value is set. Is preferably prevented from appearing. In addition,
It is needless to say that the present apparatus can be widely used not only for ships navigating on water, but also for detecting a slowly changing moving object existing in a rapidly changing noise component.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to embodiments.

[0019]

Embodiment 1 FIG. 1 is a block diagram of a moving object monitoring apparatus according to the present embodiment, FIG. 2 is a configuration diagram thereof, FIG. 3 is a conceptual diagram illustrating a detection principle in the present invention, and FIG. FIG. 5 is a process diagram schematically showing an image during the image processing process.

The moving object monitoring apparatus of this embodiment comprises a monitoring camera 1, an arithmetic processing unit 2, a display 3, and a speaker 4. The surveillance camera 1 is installed at a fixed point overlooking a port or a gateway, and supplies the acquired image to the arithmetic processing device 2. When the arithmetic processing device 2 detects a moving ship by performing image processing on the acquired image based on the method of the present invention, the result is notified to the operator through the display 3 or the speaker 4. The operator can visually confirm the movement of the ship after receiving the warning that the moving ship has been found. This device eliminates the need for the observer to keep an eye on the display at all times, so that other operations can be performed in parallel with the monitoring operation, and labor and labor can be saved in port monitoring and the like. .

The surveillance camera 1 is used with the visual field fixed in the surveillance direction. Whether or not a moving object is present in the field of view is detected using an inter-frame difference. That is, FIG.
As shown in the figure, two images with a time interval (a)
When an image difference is performed between (b) and the obtained image is binarized using an appropriate threshold to generate a binarized image (c),
Since a stationary object is located at the same position in both images, the difference is 0 and does not appear in the binarized image, but the moving object has a different position in both images and a portion corresponding to the moving portion is 1
And remains as a high-luminance area, so that it can be detected. However, when the monitoring area is the sea surface, even when there is no moving object, the wind generates a wavefront or the sun reflects on the wavefront, so a large amount of noise occurs on the screen and only the moving object is Difficult to detect.

Therefore, in the apparatus of this embodiment, noise suppression processing for suppressing the influence of sea level change is applied to the inter-frame difference method, and accumulation processing for relatively improving the detection sensitivity of a low-speed moving object such as a ship is added. As a result, a low-speed moving object is detected with a high recognition rate. This embodiment focuses on the fact that the speed of light change due to the reflection of sea surface waves and sunlight is very fast compared to the moving speed of a ship, and a bandpass filter function for detecting a low speed change is an extremely simple image. It is characterized by being realized by processing.

As shown in FIG. 1, an arithmetic processing unit 2 for executing an image operation includes a first image memory 11 for storing a picked-up image therein, two image difference units 12, 13, an image operation unit 14, and an intermediate unit. A second image memory 15 for storing images,
An image processing device 16, a determination device 17, and a notification device 18 for operating the display 3 or the speaker 4 are provided. Images captured by the monitoring camera 1 are sequentially stored in the first image memory 11 at appropriate intervals. The first image memory 11 is a memory stack in which the oldest image is deleted each time a new image is captured, and a predetermined number of images are always stored from the latest image.

The image storage interval is determined from a time interval at which a fast-changing noise component can be detected. When observing the sea surface in a port, for example, 200 to
What is necessary is just 300 milliseconds. The number of image storage frames is determined from an advantageous interval for detecting a moving object moving at a low speed in an image, and depends on the position of the surveillance camera 1. Can be.

FIG. 4 shows the processing contents of the arithmetic processing unit 2. First, in order to set an initial state, n images (here, nine images) are fetched and stored in the first image memory 11 (S1). Further, the previous moving object detection image SIt _-1 stored in the second image memory 15 as an initial setting is set to 0 (S2). On top of that, captures a new image I _t, starts the normal moving object detection process (S3). This new image I _t, is imprinted spot due to for example the hull and waves as shown in FIG. 5 (a).

The difference between frames of an image is determined by two types of combinations. First, in order to detect light noise on the sea surface,
The image I at the latest time t is stored in the first image difference device 12.
_The difference image DI ₁ (that is, DI ₁ = | I _t −I _t−1 |) is obtained by taking in _t and the image It ₋₁ one frame before, and calculating the absolute value of the difference for each pixel value. Difference image DI ₁
For example, as shown in FIG. 5B, a light spot on the sea surface is detected, but the hull portion is almost canceled and does not appear on the image.
It should be noted that the image used here is not limited to the latest image, and it is also possible to use an image substantially intermediate between the latest image and an image n away from the latest image.

At the same time, in order to reliably detect the movement of the ship, the image It at the latest time _t and the image _Itn before the nth frame are taken into the second image difference device 13 and the absolute value of the difference is determined for each pixel. To calculate the difference image DI _n (that is, DI _n
= | I _t -I _tn |) seek. (S4). Here, n is 9, for example. The difference image DI _n is, for example, as shown in FIG. 5 (c), and the front and rear ends of the moving hull remain as high brightness points in addition to the light spots on the sea surface.

Next, the image calculation unit 14 calculates the moving object detection image SI _t based on the equation below and enter the two difference images DI ₁ and DI _n (S5). SI _t = FIX {a (SI _t−1 + DI _n ) −DI ₁ } where FIX is a saturation operator, and the minimum value of a pixel value, the minimum value of 0 or less, and the minimum value of 0 The maximum value FF is taken for a value equal to or larger than the maximum value, for example, FF (hexadecimal number), and the value x is taken for an intermediate value x between them. A is a damping coefficient, which is a positive number between 0 and 1, and SIt _-1 is a second number.
Is a previously detected moving object detection image stored in the image memory 15 of FIG. The calculation is performed for each pixel position over the entire surface of the image. The attenuation coefficient a sets a window for limiting the calculation target in the time axis direction. In the above equation, (SI _t-1 + DI _n ) is multiplied by the attenuation coefficient a.
DI _n may not be multiplied by the reduction coefficient a.

According to the above equation, the attenuation coefficient a is obtained by adding the difference image DI _{n in} which the moving part of the ship and the optical noise on the water surface are visualized to the moving object detection image SI _t-1 obtained last time.
From multiplied by, those only light noise on the water surface is obtained by subtracting the difference image DI _1, which is visualized as moving object detection image SI _t, it has a repetitive operation. That is,
Moving object detecting image SI _t has include those obtained by integrating over the attenuation coefficient a in the moving object detection image obtained so far, that the accumulated image of the difference image DI ₁ from the accumulated image of the difference image DI _n by subtracting Is equivalent to window-processed.

[0030] Since the light spot generation position on the surface of the water is not determined, but it is not possible to erase the noise even in a single process cycle by subtracting the difference image DI ₁ from the difference image DI _n, accumulated by repeating the process In the meantime, noise components are equally generated and offset on the entire water surface, and the result is that noise is suppressed as a whole. In addition, since the movement of the ship is very slow as compared with the image capturing interval, the ship movement portion in the difference image DI _n between the images n distances apart largely overlaps with each image processing, while the light spot noise is random. Therefore, the contrast between the moving part of the ship and the light spot noise on the water surface becomes extremely clear by integrating these.

The image computing unit moving object detecting image SI _t obtained in 14, for example as in FIG. 5 (d), the attenuation coefficient a
The leading end portion and the trailing end portion of the moving body are overlapped for a time width determined by, and the moving trajectory of the moving body can be seen. Due to the effect of the attenuation coefficient a, newer events are darker and older events are lighter. Also, the attenuation coefficient a
Past events that have run out of the virtual window defined by are deleted from the image. Light noise on the water surface is 2
Is erased offset because it is substantially the same extent integrated into One difference image DI ₁ and DI _n. However, among the light spots on the water surface, those that were not offset by the two difference images slightly remain in the image. Moving object detecting image SI _t is stored in the second image memory 15 (S6), it is used for the next operation.

Next, the moving state of the generated mobile binarized image darker in brighter still region movement area from the moving object detecting image SI _t facilitate determination visualized. First,
By creating a density histogram of the latest captured image I _t to determine the appropriate binarization threshold lightness distribution on the screen (S
7). Mobile moving portion in the moving object detection image SI _t is displayed as brightness difference of the background portion and the moving body is accumulated on multiplied by the predetermined attenuation rate.

For example, under a normal light condition in which sunlight is radiated from behind the camera, an image having a small spread of the density histogram is obtained because the light beam is settled on the entire screen.
In this case, it is necessary to set a lower binarization threshold of the moving object detection image SI _t for background and brightness difference of the moving object is not so large. On the other hand, in the case of a backlight condition in which the sun irradiates from the front of the camera, the sunlight reflected on the wavefront becomes a high-brightness light spot that changes at a high speed and enters the camera, so that the density histogram on the screen is greatly expanded. . In this case, 2
If the binarization threshold is raised, fine noise is picked up and advanced image processing is required.However, since the brightness difference between the background and the moving object increases, it is possible to easily remove the fine noise by increasing the binarization threshold. it can. Therefore, to determine a binarization threshold based on the spread of the density histogram of the captured image I _t. Note that the captured image used to determine the binarization threshold may not be the latest one.

The brightness value for each pixel in the moving object detection image SI _t is determined whether greater or not than the binarization threshold to generate a binary image BI _t displaying 0,1 based on the result ( S8). The binarized image BI _t, movement area still region is darker expressed brightly. Moving area portion in the binary image BI _t has the length proportional to the width and the moving speed of the moving body. The binarized image BI _t, there is a case where the light spot noise included. In the mobile monitoring device of the present embodiment, by determining the binarization threshold based on the spread of the density histogram of the captured image I _t, thereby enabling appropriate binarization any case frontlit, backlit .

Next, the binarized image BI _t numbering detects continuous area by the labeling processing (S9). Then, the number of pixels in the region having the same number is counted and set as a value representative of the area S of the continuous region (S10). The noise component has a small area S, and the area S of the moving area of the moving object is larger than the noise. Then, in order to determine the moving object, each area S is compared with a predetermined threshold value S _th (S
11).

The region area S is determined that the mobile is greater than the threshold S _th is present, informs the operator via the display 3 or the speaker 4 by the notification device 18 (S12). When the operator receives the notification, the operator can check the moving object again by the photographed image displayed on the monitor or directly, and can perform necessary work. If the area S does not reach the threshold value S _th , it is determined that there is no moving object, the process returns to the image capturing step (S3), and the same process is repeated. As described above, since there is no need to constantly monitor if the mobile monitoring apparatus of the present embodiment is present, the operator can perform monitoring work in parallel with other work.

[0037]

[Embodiment 2] FIG. 6 is a flowchart of image processing in the moving object monitoring apparatus of the present embodiment, and FIG. 7 is a process diagram schematically showing an image in an image processing process.

The moving object monitoring apparatus of the present embodiment is different from the first embodiment only in the image processing process. Therefore, the different parts will be mainly described with reference to FIGS. First, in preparation for arithmetic processing, n images are fetched and stored in the first image memory 1.
1 is stored. In addition, as the initial setting,
_Dnt-1 and ΣD1t _-1 are each set to 0 (S21). On top of that, captures a new image I _t, starts the normal moving object detection process (S22). This new image I _t will image as shown in (a) of FIG. Then, the latest image in the first image difference device 12 I _t and the previous frame image I
difference image DI ₁ calculates the absolute value of the difference for the _{t-1 (i.e., DI 1 = | I t -I} t-1 |) determined. Difference image D
I ₁ is as shown in FIG. 7 of (b).

Further, the second image difference device 13 to the latest image I _t and n frames before the image I _tn calculates the absolute value of the difference the difference image DI _n (i.e., DI _n = | I _t -I
_tn |). (S23). This difference image DI _n is as shown in FIG. Next, the image processing device 14
The two difference images DI ₁ and DI _n are input and 2
Two integrated images ΣD _1t and ΣD _nt are calculated (S24). ΣD _nt = DI _n + aΣD _nt-1 ΣD _1t = DI ₁ + aΣD _1t-1 where a is an attenuation coefficient.

As shown in FIG. 7D, the integrated image ΣD _1t is a screen in which the hull is white and the light spots on the water surface are spread evenly. The new light spot is darker and the old light spot is thinner. Further, the accumulated image .SIGMA.D _nt is a screen dark area having a width in the moving direction other than the light spot on the water surface at the tip portion and the rear portion of the hull, as shown in FIG. 7 (e) appeared. The width of the area that appears before and after the hull is affected by the speed of movement and the damping coefficient. Also, the moving part of the hull is newer and darker and older and thinner.

The image calculation device 14 further calculates the integrated image ΣD
ΣD _1t is subtracted from _nt to detect the moving object detection image SI _t (SI _t = Σ
D _nt -ΣD _1t) calculates a (S25). Moving object detecting image SI _t thus obtained, as shown in FIG. 7 (f),
This is almost the same as the moving object detection image obtained by the moving object monitoring device of the first embodiment, and a moving object moving at a low speed can be detected by offsetting the light spot noise that changes at a high speed. The subsequent processing is the same as in Example 1, to determine the appropriate binarization threshold from the density histogram of the captured image I _t, binary image by binarizing the moving object detection image SI _t Based on this B
To generate the I _t (S26). Binary image BI _t Figure 7
(G).

Next, the binarized image BI _t is labeling (S27), the area S _i of each region independent appearing in the image is determined (S28). Furthermore, compared with the predetermined threshold value S _th each region area S _i (S29), determines that the mobile if the threshold S _{t h} a larger region area S _i is present, the operator notification by the notification device 18 Yes (S3
0). If there is no area S _i that has reached the threshold value S _th , it is determined that there is no moving object, the process returns to the image capturing step (S22), and the same processing is repeated thereafter.

[0043]

As described above, the moving object monitoring method and apparatus according to the present invention can detect a relatively slow moving object from an image in which a large amount of rapidly changing noise is generated. It can be used to automatically detect a ship moving on the water surface, such as a gate and a gate monitor, and can realize labor-saving and labor-saving monitoring work.

[Brief description of the drawings]

FIG. 1 is a block diagram of a mobile monitoring apparatus according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of the present embodiment.

FIG. 3 is a conceptual diagram illustrating a detection principle according to the present invention.

FIG. 4 is a flowchart of image processing in the present embodiment.

FIG. 5 is a process chart schematically showing an image in an image processing step in the present embodiment.

FIG. 6 is a flowchart of image processing in a second embodiment of the present invention.

FIG. 7 is a process diagram schematically illustrating an image in an image processing process in the present embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 surveillance camera 2 arithmetic processing device 3 display 4 speaker 11 first image memory 12, 13 image difference device 14 image calculation device 15 second image memory 16 image processing device 17 determination device 18 notification device

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Nobuo Nagai 1-1, Kawasaki-cho, Akashi-shi, Hyogo Prefecture Kawashige Techno Service Co., Ltd. F-term (reference) in Kanto Works Co., Ltd. EA07 FC01 FC05 FC13 FF06 GA04 HA18 HA26 5L096 AA06 BA02 CA02 DA03 EA43 FA35 FA59 GA08 HA03

Claims (6)

[Claims] 1. A captured image of a target area is captured at a predetermined cycle, and each time a captured image is captured, an absolute value of a difference between adjacent captured images is calculated as a first difference image, and captured at a predetermined interval. An absolute value of the difference between the images is taken as a second difference image, and a combined image obtained by adding the first difference image to the same predetermined number of images is subtracted from the sum of the second difference image and a predetermined number of images. Forming a binary image using a predetermined threshold value to obtain a binary image, and notifying when a region larger than a predetermined area is detected from a video of the binary image. Mobile monitoring method. 2. A photographic image of a target area is fetched at a predetermined cycle, and a predetermined number of photographic images of the target area are sequentially stored at regular intervals from the latest one, and the difference between the latest photographic image and an adjacent photographic image is calculated. A second difference value which is an absolute value of a difference between the first difference image which is an absolute value and a shot image separated by a predetermined distance from the latest shot image.
Forming a difference image, subtracting the first difference image from the image obtained by multiplying the previously obtained intermediate image and the second difference image by an attenuation coefficient, and forming a composite image as an intermediate image;
Thereafter, the same process is repeated, and during this time, the composite image is binarized using a predetermined threshold to obtain a binarized image, and a notification is made when an area larger than a predetermined area is detected from the image of the binarized image. A method for monitoring a moving object, comprising: 3. The moving object monitoring method according to claim 1, wherein the binarization threshold is determined based on a histogram relating to brightness in a captured image of the target area. 4. An image capturing apparatus, an image processing apparatus, an image memory, and an alarm device, wherein the image capturing apparatus captures an image of a target area at a predetermined cycle, and the image memory stores a captured image memory, an intermediate image memory, and a composite image memory. A predetermined number of equally-spaced images acquired by the imaging device are sequentially stored in the captured image memory from the latest one, and the image processing device stores the image at the absolute value of the difference between the latest image and the adjacent image. Forming a second difference image having an absolute value of a difference between the one difference image and an image separated by a predetermined distance, adding the intermediate image recorded in the intermediate image memory and the second difference image, and multiplying by an attenuation coefficient; Subtracting the first difference image from the image to form a composite image,
The composite image is stored in the intermediate image memory, and when a binarized image obtained by binarizing the composite image is detected to detect a region larger than a predetermined area, an alarm is issued via the alarm device. A moving object monitoring device characterized in that: 5. The moving object monitoring apparatus according to claim 4, wherein the binarization threshold is determined based on a histogram relating to brightness in an image of the target area. 6. The moving object monitoring apparatus according to claim 4, wherein the intermediate image is saturated at an upper limit value or a lower limit value for each pixel.